# Investigating Disentanglement in a Phoneme-level Speech Codec for Prosody Modeling

arXiv: https://arxiv.org/abs/2409.08664 · HTML: https://mlsanity.com/paper/2409.08664/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 25 of its 31 references that are in the map.
- First posted on arXiv 13 Sep 2024.

## Builds on (most cited first, 25 of 25)

- [Neural Discrete Representation Learning](https://mlsanity.com/paper/1711.00937/) (2017)
- [Robust Speech Recognition via Large-Scale Weak Supervision](https://mlsanity.com/paper/2212.04356/) (2022)
- [Conformer: Convolution-augmented Transformer for Speech Recognition](https://mlsanity.com/paper/2005.08100/) (2020)
- [HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesis](https://mlsanity.com/paper/2010.05646/) (2020)
- [SoundStream: An End-to-End Neural Audio Codec](https://mlsanity.com/paper/2107.03312/) (2021)
- [Neural Codec Language Models are Zero-Shot Text to Speech Synthesizers](https://mlsanity.com/paper/2301.02111/) (2023)
- [FastSpeech 2: Fast and High-Quality End-to-End Text to Speech](https://mlsanity.com/paper/2006.04558/) (2020)
- [High Fidelity Neural Audio Compression](https://mlsanity.com/paper/2210.13438/) (2022)
- [AudioLM: a Language Modeling Approach to Audio Generation](https://mlsanity.com/paper/2209.03143/) (2022)
- [Conditional Variational Autoencoder with Adversarial Learning for End-to-End Text-to-Speech](https://mlsanity.com/paper/2106.06103/) (2021)
- [Generalized End-to-End Loss for Speaker Verification](https://mlsanity.com/paper/1710.10467/) (2017)
- [Style Tokens: Unsupervised Style Modeling, Control and Transfer in End-to-End Speech Synthesis](https://mlsanity.com/paper/1803.09017/) (2018)
- [NaturalSpeech 2: Latent Diffusion Models are Natural and Zero-Shot Speech and Singing Synthesizers](https://mlsanity.com/paper/2304.09116/) (2023)
- [NaturalSpeech: End-to-End Text to Speech Synthesis with Human-Level Quality](https://mlsanity.com/paper/2205.04421/) (2022)
- [SoundStorm: Efficient Parallel Audio Generation](https://mlsanity.com/paper/2305.09636/) (2023)
- [Hierarchical Generative Modeling for Controllable Speech Synthesis](https://mlsanity.com/paper/1810.07217/) (2018)
- [Meta-StyleSpeech : Multi-Speaker Adaptive Text-to-Speech Generation](https://mlsanity.com/paper/2106.03153/) (2021)
- [Non-Attentive Tacotron: Robust and Controllable Neural TTS Synthesis Including Unsupervised Duration Modeling](https://mlsanity.com/paper/2010.04301/) (2020)
- [Robust and fine-grained prosody control of end-to-end speech synthesis](https://mlsanity.com/paper/1811.02122/) (2018)
- [Fully-hierarchical fine-grained prosody modeling for interpretable speech synthesis](https://mlsanity.com/paper/2002.03785/) (2020)
- [Mega-TTS 2: Boosting Prompting Mechanisms for Zero-Shot Speech Synthesis](https://mlsanity.com/paper/2307.07218/) (2023)
- [Generating diverse and natural text-to-speech samples using a quantized fine-grained VAE and auto-regressive prosody prior](https://mlsanity.com/paper/2002.03788/) (2020)
- [StyleTTS: A Style-Based Generative Model for Natural and Diverse Text-to-Speech Synthesis](https://mlsanity.com/paper/2205.15439/) (2022)
- [Exploring Disentanglement with Multilingual and Monolingual VQ-VAE](https://mlsanity.com/paper/2105.01573/) (2021)
- [Controllable and Lossless Non-Autoregressive End-to-End Text-to-Speech](https://mlsanity.com/paper/2207.06088/) (2022)

## Built on by (most cited first, 1 of 1)

- [Recent Advances in Discrete Speech Tokens: A Review](https://mlsanity.com/paper/2502.06490/) (2025)

## Abstract (arXiv)

Most of the prevalent approaches in speech prosody modeling rely on learning global style representations in a continuous latent space which encode and transfer the attributes of reference speech. However, recent work on neural codecs which are based on Residual Vector Quantization (RVQ) already shows great potential offering distinct advantages. We investigate the prosody modeling capabilities of the discrete space of such an RVQ-VAE model, modifying it to operate on the phoneme-level. We condition both the encoder and decoder of the model on linguistic representations and apply a global speaker embedding in order to factor out both phonetic and speaker information. We conduct an extensive set of investigations based on subjective experiments and objective measures to show that the phoneme-level discrete latent representations obtained this way achieves a high degree of disentanglement, capturing fine-grained prosodic information that is robust and transferable. The latent space turns out to have interpretable structure with its principal components corresponding to pitch and energy.